Federico Vasile

I'm a Postdoc at Istituto Italiano di Tecnologia, in the Humanoid Sensing and Perception (HSP) group led by Lorenzo Natale. During my PhD journey, I spent six wonderful months visiting UC San Diego, where I worked with Prof. Xiaolong Wang.

My research lies at the intersection of Computer Vision, Robotics and Prosthetics. I develop algorithms for autonomous hand prosthesis control from visual input, aiming to improve the quality of life of amputees.
Check out my research statement here.

Email / CV / Scholar / Twitter / Github

Research

Papers sorted by recency. Videos play automatically.

	Gaussian-Augmented Physics Simulation and System Identification with Complex Colliders Federico Vasile, Ri-Zhao Qiu, Lorenzo Natale, Xiaolong Wang, NeurIPS, 2025 project website / paper / code A simulation-rendering framework for realistic collisions with arbitrary shapes.
	HannesImitation: Grasping with the Hannes Prosthetic Hand via Imitation Learning Carlo Alessi, Federico Vasile, Federico Ceola, Giulia Pasquale, Nicolò Boccardo, Lorenzo Natale IROS, 2025 project website / paper / code Grasping objects with the Hannes Prosthesis via Imitation Learning from an eye-in-hand camera.
	Continuous Wrist Control on the Hannes Prosthesis: a Vision-based Shared Autonomy Framework Federico Vasile, Elisa Maiettini, Giulia Pasquale, Nicolò Boccardo, Lorenzo Natale ICRA, 2025 project website / paper / code Segmentation and visual servoing for wrist control using an eye-in-hand camera.
	Bring Your Own Grasp Generator: Leveraging Robot Grasp Generation for Prosthetic Grasping Giuseppe Stracquadanio, Federico Vasile, Elisa Maiettini, Nicolò Boccardo, Lorenzo Natale ICRA, 2025 project website / paper / code Bridging the gap between robotic and prosthetic grasping through depth estimation, robotic grasp generation and visual odometry.
	Grasp Pre-shape Selection by Synthetic Training: Eye-in-hand Shared Control on the Hannes Prosthesis Federico Vasile, Elisa Maiettini, Giulia Pasquale, Nicolò Boccardo, Lorenzo Natale IROS, 2022 project website / paper / code (simulation) / code (experiments) A synthetic data generation framework for grasp pre-shape prediction with object-part granularity.

Borrowed from Jon Barron's source code.